100%(14)14 out of 14 people found this document helpful
This preview shows page 5 - 10 out of 11 pages.
999590807060504030201051Standardized ResidualPercentNormal Probability Plot(response is Price ($))605958575655545352210-1-2Fitted ValueStandardized ResidualVersus Fits(response is Score)
3210-1-2-3999590807060504030201051Standardized ResidualPercentNormal Probability Plot(response is Score)7.06.56.05.55.04.54.0656055504540Weight (oz.)ScoreScatterplot of Score vs Weight (oz.)
16151413121110656055504540MegapixelsScoreScatterplot of Score vs Megapixels40035030025020015010070656055504540Price ($)ScoreScatterplot of Score vs Price ($)ResultsThe results show that there is little correlation between a cameras megapixels and weight affecting the price and score of the camera. R2 is a measure of how close the data is to the fitted
line of regression, or how much the model explains or doesn’t explain the variability of the response data around its mean. The resulting r2for the regression analysis for price ($) versus megapixels and weight was 16.68%. The r2for the regression analysis for score versus megapixels and weight resulted in 8.41%. These regression scores show that very little of the variability in price and score are explained by the predictors (weight and megapixels). These values show that price and score are influenced very little to almost none by megapixels and weight. Also, the Standard error of regression (S) shows that for price versus Megapixels and weight, the average distance from the line of regression is 78.54. For Score versus megapixels and weight, the average distance from the line of regression is 6.65. S shows how wrong, on average, the regression model is using the response variable. Smaller values are better because they are closer to the fitted line.Price seems to be the best overall predictor of score because the scatterplots show that there is little deviation from the line of regression, resulting in a much higher r2value than any other variable assessed in the study. Score is influenced by price, which can be seen in the regression analysis. It shows very little departure from the line of regression. Regression Analysis: Score versus Price ($) Analysis of VarianceSource DF Adj SS Adj MS F-Value P-ValueRegression 1 311.66 311.660 23.80 0.000Price ($) 1 311.66 311.660 23.80 0.000Error 11 144.03 13.094Lack-of-Fit 6 104.78 17.464 2.22 0.199Pure Error 5 39.25 7.850Total 12 455.69Model SummaryS R-sq R-sq(adj) R-sq(pred)3.61854 68.39% 65.52% 53.13%CoefficientsTerm Coef SE Coef T-Value P-Value VIFConstant 47.29 2.57 18.38 0.000Price ($) 0.0665 0.0136 4.88 0.000 1.00Regression EquationScore = 47.29 + 0.0665 Price ($)Fits and Diagnostics for Unusual ObservationsObs Score Fit Resid Std Resid13 46.00 53.27 -7.27 -2.21 RR Large residual
Regression Analysis: Score versus Price ($) Analysis of VarianceSource DF Adj SS Adj MS F-Value P-ValueRegression 1 311.66 311.660 23.80 0.000Price ($) 1 311.66 311.660